HEP-SCORE deployment TF meeting #29

Wednesday 1 Jun 2022, 16:00 → 17:00 Europe/Zurich

Zoomland

Helge Meinhard (CERN)

Draft minutes

16:00 → 16:05 Welcome, note-taking, notes from previous meeting

minutes approved of previous meeting [PDF]

16:05 → 16:20 Progress on workloads

Speaker: Domenico Giordano (CERN)

atlas reco Validation.png

memory usage from leftover processes

• Atlas reco WL
  • Validation ongoing. Good reproducibility of scores so far (see plot )
  • Soon a script E will be available with this WL. (the last script!)
• Alice
  • scriptDv2 has been distributed (includes Alice gen-sim + IGWN)
    • Caveat: we noticed for the Alice gen-sim-digi that some O2-sim-* processes could remain pending after the singularity ends.
      • Effect: occupy memory (see plot)
      • Request for sites: please check that after the execution of script Dv2 no spurious processes remain
        • ps -ef | grep alice
        • ps -ef | grep O2-sim
  • Alice gen-sim-digi-reco container has been prepared and succeeds in the standalone mode
    • Solved the past problem of the missing libraries, forcing the cvmfs snapshot area
    • But: a number of software improvements are expected to be included in the Alice software in the near time.
      • Benefit: improve overall performance of the reco stage
    • Proposal: wait for these improvements before creating an official Alice gen-sim-digi-reco
      • Consequence: the HEPscore candidates will be studied only with Alice gen-sim

16:20 → 16:30 Status of benchmarking campaigns

Speaker: Gonzalo Menendez Borge (CERN)

2022-06-01_TF_Measurement_Collection_Update@TF.pdf

2022-06-01_TF_Measurement_Collection_Update@TF.pptx

CPUs_missing_entries-2022-06-01.png

16:30 → 16:35 Any other business

• Workshop in September: Date, venue